AA80E 8-speed

[Jim Blackwood]

Just an update on some off forum discussion about possibly using two of the LED lines for solenoid outputs. I don't think the LED drivers will handle the current. They can be upgraded to 2 Amp transistors (ZTX651) which should be OK at about an amp and a half max with a half amp cushion both at the transistor and at the board traces. Personally I don't have a problem with off the board drivers, they can be attached to the case end plate and connected with stranded leads, or a small proto board attached somewhere to the main board and that allows the use of a 3-5 Amp driver which should improve reliability, particularly with solenoid drivers. I'll be looking into mounting soon, but I wanted to bring this up here and maybe get Lance's input and possibly others. Particularly Lance as he has to make the code changes. I'd sort of like to keep the LED drivers but have no problem with giving up one of them for another purpose as this only means 1st and 8th gears are blanked out. Two of them means the gear indicator no longer provides much useful information.

We do have some excess capacity in the shifter inputs, specifically redundancy in the Park/Neutral lines. We also have the non-CAN input line, and one of these two could be used for a switchable manual mode switch. It occurs to me that there may be times when the driver does not want the controller to time out and revert to auto mode, such as when towing. Probably the redundant shifter input would be the best choice for this. Neutral is EGT3, Park is EGT4. I would pick EGT4 as the manual mode select input unless anyone has a better recommendation.

Jim

[Bernard Fife]

Jim,

I am not ignoring this thread, I am just swamped at the moment on a number of fronts, and I want to make sure I am able to fully concentrate on the information here when I do tackle it.

BTW, the manual mode doesn't 'time out'. Once you select manual mode (using a shift button) the controller stays in manual mode until the rpms limits are exceeded, or the VSS drops below the minimum speed (and only if the user has enabled these and set the controller to go to auto mode if they are exceeded). So you probably don't need to burn up an input just to keep the trans in manual for towing.

Lance.

[Jim Blackwood]

Thanks Lance, no need to feel rushed, we're still both some ways from testing.

I don't have a real feel yet for how the manual/auto changeover will work in actual use, so just trying to provide for possible future needs. We seem to be good on outputs and inputs at this point, with one or two possible spares. I'm guessing there might be additional needs for a standalone controller but neither of us is going that route and I haven't looked at what would be required or how those needs could be met within what we now have.

As it now stands we can use the 1st LED circuit as a spare if needed, the park input as a spare which could be used if there is a reason that a manual/auto switch is desired, the non-CAN input and I think that's about it. Standalone there would also be the CAN circuits I suppose, and maybe that combination would be enough since we are using both the input and output speed sensors.

I've gotten back on the circuit build but it's slow going. I have to double check some of the circuit schematics. Also owe Matt a phone call later today. I know what you mean about being swamped.

Later dude.
Jim

[Jim Blackwood]

Just a short update.
To be completely honest, I've been putting off the controls til the last, and so I'll be running out of other things to do before too much longer. I did get the connector body for the gear selector switch, it is available from Toyota for about ten bucks. If anyone needs the body for the main connector I have 3 spares.

Can't say for sure when I'll get back on the controller build, still need to verify those 8 or 9 circuits and I think some redesign there could be needed. If I were to guess I'd say perhaps by Thanksgiving. I do need a small inexpensive thermocouple amp to provide a 0-5vdc output to the MS controller for my MAT input.

Other than that and a few connectors I'm now well into the plumbing and still need to build the shifters. I think everything else is done though.

Jim

[Jim Blackwood]

FINALLY, I think I've got it. This time for sure, right? Please have a look and report any errors you notice if you will. I'm about brain numb at this point. The mostly verified spreadsheet should be attached.

Anyway, I went back to Lance's original recommended circuit assignments and deviated from that only a little. Here the build instructions show 2 jumper wires and I've added a third, completing the loop and allowing the use of PEO on VR4.

Circuit change #1: Moved the thermister input -ATF Temp- to PEO/VR4 (Input only allowed).
Circuit change #2: Used the AD2/GPI3 circuit with an external driver for Shift Trigger solenoid SR. This is a 12V+ output.
Circuit change #3: Used ADO/EGT4 with an external driver for TCC/Reverse sloenoid SL. This is a 12V+ output.
I deleted the redundant park and neutral input, using the interlock Park/Neutral switch contact coming in on the Neutral circuit AD3/EGT3.

Provided these changes are acceptable, one question remains regarding the VR circuits. Well, two really.

Question #1: For a thermistor (temperature variable resistor) will the standard circuitry of VR4 work, is a pullup resistor needed, and if so is 330 ohms the correct value?
Question #2: For the Hall Effect speed sensors NP and SP2, will the standard circuitry of VR3 and VR1 work by simply adding a pullup resistor of 330 ohms?

Thanks for all of the help, with those two questions answered I can give final approval to the spreadsheet if there are no other errors and get on with the build.
In the end the results are deceptively simple. It was the getting there that was difficult, and if in the end I've managed to make it look like nothing was changed while everything works correctly then I'm satisfied and very happy with the results.

I think the documentation is comprehensive. Feel free to use it in any way that complies with the ULA and rules of this site.

Jim

[Bernard Fife]

Jim,

Circuit change #1: VR4 will work, but PE0 won't. It is digital only, as well as input only, so it doesn't have the ADC function necessary to turn a variable resistance (and voltage) into a temperature. Only circuit pins of the form ADx (sometimes aka. "AD0x" or PAD0x) have the ADC (analog to digital converter) function.

Circuit change #2: the AD2/GPI3 (AD2 is the temperature sensor by default in the code) circuit can be used as an output. It will require an external circuit (or some clever adaptation of the existing circuit with some add-ons).

Circuit change #3: the ADO/EGT4 (AD0 is Input 1 by default) this is the same situation as change #2. It can be used, but requires some mods.

Question #1: For a thermistor (temperature variable resistor) will the standard circuitry of VR4 work, is a pullup resistor needed, and if so is 330 ohms the correct value?

The VR circuitry will need modification from its VR use, you would build it like VR2 in step J here: http://www.msgpio.com/manuals/mshift/assembly.html Also see: http://www.msgpio.com/manuals/vr.htm for component equivalents.

Question #2: For the Hall Effect speed sensors NP and SP2, will the standard circuitry of VR3 and VR1 work by simply adding a pullup resistor of 330 ohms?

Yes, the standard circuits should work for this.

So you might want to swap PE0 with another digital input (like the 4WD/speedo switch) and swap that pin with one of the two ADC pins you have specified. Leave the temp sensor on AD2.

I haven't looked at all the implications of such moves, though.

Lance.

[Jim Blackwood]

Thanks Lance, good stuff.

I put the pressure varistor back on AD2/GPI3 where it was to start with.

I moved the Shift Trigger solenoid SR to AD4/EGT2. Both EGT2 and AD0/EGT4 will need an off board TIP120 to provide a 12V+ output with a flyback diode to drive the two non-PWM solenoids. I think there might be room for them on one side of the Ampseal connector.

At PEO/VR4 I put a momentary pushbutton to engage "Auto" mode. I thought it might not be too difficult to use that button while in auto mode as a downshift button without leaving Auto mode, sort of like what some of the new cars have.

I think that keeps the number of circuit changes the same as before, so it retains the same minimum impact on the legacy designs.

I think the only other thing I did differently was to reserve AD5/EGT1 (non-CAN MAF, TPS, etc) for my planned "Clutch" function at some future date. I think it is a workable idea with certain coded-in limitations and something I'd like to experiment with after everything else is proven out. What I have in mind there is basically to deactivate all solenoids and then base re-engagement on system inputs with upper limits to prevent transmission damage. It is most emphatically NOT intended for use as a launch control.

I entered the changes to the spreadsheet, which awaits only your approval at this point. I believe it is now correct in every particular. It should be attached.

Jim

[Jim Blackwood]

Update: The engine now runs, under MS-II. Still sorting out a few issues there like leaks and charging. The car does move under it's own power with no electrical power to the transmission, but not particularly well. Limp mode. I will check the fluid again, it may need topped up a second time.

I plan to get back on the controller in another week or two, finish that up and begin the install. Paddle shift will come later. So will MS-III.

Jim

[Jim Blackwood]

OK the board is now finished. On the pull-up resistors for the VR circuits I tied into the C(a) cap location as recommended in the GPIO build instructions. Otherwise all in accordance with the spreadsheet with the following variations:

Pull-up resistors for VR2 and VR4 were 330 ohm simply because I already had them installed on the board.
GPI3 is a 2.49K for the same reason.
Power for the two extra TIP120's came off the top side of AMP pin one as being more convenient.
I would also suggest maybe turning the PCB over if the same one is used as it would be easier to solder with the traces on the outside.

Ready for clean-up and a shot of lacquer to seal it and in the box it goes. A little sooner than I expected. Next, the wiring harness. I got Matt to sell me long wires in rainbow colors so I will probably add the wire colors I assign to the spreadsheet.

Jim

[Jim Blackwood]

Final Spreadsheet with AMP wiring harness assignments.

Jim